Fixed log-loss latex formula in Chapter 5

CamDavidsonPilon / Probabilistic-Programming-and-Bayesian-Methods-for-Hackers

aka "Bayesian Methods for Hackers": An introduction to Bayesian methods + probabilistic programming with a computation/understanding-first, mathematics-second point of view. All in pure Python ;)

http://camdavidsonpilon.github.io/Probabilistic-Programming-and-Bayesian-Methods-for-Hackers/

MIT License

26.51k stars 7.84k forks source link

Fixed log-loss latex formula in Chapter 5 #438

Closed chdamianos closed 5 years ago

chdamianos commented 5 years ago

I think the log-loss definition in Chapter 5 is wrong. The logarithm should be applied to the estimated probabilities. I swapped around the \\hat{} to correct.

CamDavidsonPilon commented 5 years ago

A larger problem is that I have a subtraction sign between the terms, not an addition sign.

chdamianos commented 5 years ago

I think the formula is correct. You have minus on both terms. https://en.wikipedia.org/wiki/Loss_functions_for_classification#Cross_entropy_loss_(Log_Loss)

CamDavidsonPilon commented 5 years ago

oh right okay

CamDavidsonPilon commented 5 years ago

w.r.t to these changes 👍